知搜-全球专利检索

  1. Login
  2. Sign Up

Search Results

14 records (took 0.017 seconds)

    DATA SECURITY
    2.
    发明申请
    DATA SECURITY 有权

    公开(公告)号：US20210365581A1

    公开(公告)日：2021-11-25

    申请号：US17444245

    申请日：2021-08-02

    Applicant: Palantir Technologies Inc.

    Inventor： Samuel Szuflita ,  Alice Yu ,  Emily Wang ,  Hao Dang ,  Megha Arora ,  Nicholas Gates ,  Samuel Rogerson

    IPC: G06F21/62 ,  G06F16/36 ,  G06F16/903 ,  G06F21/60 ,  G06F16/901

    Abstract: A computer system is configured to receiving a data set from a data provider and automatically save the data set in a quarantine database where copying, moving, and sharing of the data set are restricted until the data set is released by a data provider. The data set is parsed to find and mark portions with potentially sensitive information. At least those parts are reviewed by a data governor, who can confirm, add, edit, or remove markers. Those parts can be visually indicated to the data governor, along with a preview of, metadata about, and analysis of the data set. After reviewing at least the automatically marked portions, the data governor can release the data set to a non-quarantine database where another user can use the data set. The user is restricted from accessing the quarantine database.

    FRAMEWORK FOR EVALUATION OF COMPUTER-BASED MODELS
    5.
    发明申请
    FRAMEWORK FOR EVALUATION OF COMPUTER-BASED MODELS 有权

    公开(公告)号：US20240420258A1

    公开(公告)日：2024-12-19

    申请号：US18349738

    申请日：2023-07-10

    Applicant: Palantir Technologies Inc.

    Inventor： Arnav Jagasia ,  Megha Arora ,  Sophie Scannell ,  Emily Su ,  Meghan Nayan ,  Angela McNeal ,  David Lisuk ,  Jim Inoue ,  Justin Ngai

    IPC: G06Q50/16

    Abstract: Computer-implemented systems and methods are disclosed, including for evaluation of computer-based models in a management framework. A computer-implemented method may include, for example, receiving one or more inputs including requesting to add an evaluation configuration to a defined modeling objective, specifying at least a first evaluation data set for the evaluation configuration, specifying at least a first evaluation library for the evaluation configuration, and specifying at least a first subset definition for the evaluation configuration. A computer-implemented method may in response to the one or more user inputs include: creating, storing, and/or updating the evaluation configuration. A computer-implemented method may include evaluating, based on the evaluation configuration, the one or more models associated with the defined modeling objective.

    OPERATIONALISING FEEDBACK LOOPS
    6.
    发明公开
    OPERATIONALISING FEEDBACK LOOPS 审中-公开

    公开(公告)号：US20240354646A1

    公开(公告)日：2024-10-24

    申请号：US18639394

    申请日：2024-04-18

    Applicant: Palantir Technologies Inc.

    Inventor： Alexander Edwards ,  Anirvan Mukherjee ,  David Kebudi ,  Megha Arora ,  Sriram Krishnan ,  Max-Philipp Schrader ,  Jessica Winssinger ,  Philipp Hoefer

    IPC: G06N20/00

    CPC classification number: G06N20/00

    Abstract: Disclosed herein is a method of providing feedback to a machine learning model. The method includes allowing a user to observe an output of a trained machine learning model; allowing the user to input feedback to the machine learning model based on the output, wherein the feedback is on at least one of a model level or on a training dataset level; and incorporating the feedback into the machine learning model to improve the machine learning model, wherein the method is performed using one or more processors. Disclosed herein are one or more computer-readable storage media including computer executable instructions which when executed by the one or more processors cause the one or more processors to perform the method. Disclosed herein is a computer system which includes one or more processors and one or more computer-readable storage media which include computer executable instructions which when executed by the one or more processors cause the one or more processors to perform the method.

    MULTI-PARTY MODEL TRAINING
    7.
    发明公开
    MULTI-PARTY MODEL TRAINING 审中-公开

    公开(公告)号：US20230359905A1

    公开(公告)日：2023-11-09

    申请号：US18307653

    申请日：2023-04-26

    Applicant: Palantir Technologies Inc.

    Inventor： Arnav Jagasia ,  Emanuel Nastasa ,  Megha Arora

    IPC: G06N3/098 ,  G06N3/0442

    CPC classification number: G06N3/098 ,  G06N3/0442

    Abstract: Methods and systems for collaboration between two or more parties to generate a model based on data from the parties without data to the another party. A system includes an aggregation system having computer storage devices configured to store a model and a plurality versions of an aggregated model, and instructions, and one or more processors configured to execute the plurality of computer readable instructions to, iteratively, receive a first updated model from a first system controlled by a first party and a second updated model from a second system controlled by a second party, determine changes from the first updated model and the second updated model to include in an aggregated model, communicate a version of the aggregated model to the first system and a version of the aggregated model to the second system, and store a final aggregated model.

    AGENT OPS FRAMEWORK
    8.
    发明申请
    AGENT OPS FRAMEWORK 有权

    公开(公告)号：US20250110786A1

    公开(公告)日：2025-04-03

    申请号：US18794966

    申请日：2024-08-05

    Applicant: Palantir Technologies Inc.

    Inventor： Max-Philipp Schrader ,  Sriram Krishnan ,  Megha Arora ,  Pavlo Tyshevskyi ,  Johannes Beil ,  Alexander Edwards ,  David Kebudi ,  Montgomery Evans ,  Anirvan Mukherjee

    IPC: G06F9/50 ,  G06F9/54

    Abstract: Computer-implemented systems and methods are disclosed, including systems and methods for automatically solving computational tasks or problems. A computer-implemented method may include: providing an agent service that utilizes a plurality of agents to process one or more tasks; receiving, by a first agent, a request to handle a first task; obtaining, by the first agent, a first accessory to handle the first task; assigning, by the first agent, at least a portion of the first task to a second agent; sharing, by the first agent, the first accessory to the second agent; and processing, by the second agent, at least the portion of the first task using the first accessory to generate a processing result.

Patent Agency Ranking